What Personality Research Says About Apologies

What can personality research tell you after one specific mistake?

After one mistake, personality research can suggest questions about tendencies, but it cannot explain why someone apologized, establish whether the words were sincere, or decide what the recipient should do. Begin with what happened, its effect, your part, and any possible repair. Treat personality as a tentative way to reflect on patterns across comparable situations, not as a verdict on this one. Imagine, hypothetically, that you promised to call someone and forgot. “I’m sorry” is the apology content; whether the other person accepts it is a separate judgment; remembering or missing the next agreed call is later behavior. A personality measure may describe how readily people generally report apologizing. In “Please accept my sincere and humble apologies: The HEXACO model of personality and propensity to apologize,” reported apology proclivity was associated with measured traits in adult samples. That finding concerns an average association, not the motive behind an apology. “An Exploration of the Structure of Effective Apologies” examines how people rated apology components in trust-violation scenarios; those ratings likewise do not certify sincerity or show that a relationship was repaired. So address the event first, then consider a pattern: address the impact, then consider whether similar responses recur.

Sources: Please accept my sincere and humble apologies: The HEXACO model of personality and propensity to apologize; An Exploration of the Structure of Effective Apologies

What does a personality study actually measure when it studies apologizing?

When researchers study a personality link with apologizing, they often measure a broad reported tendency: how ready someone says they generally are to apologize for a transgression. This is called proclivity to apologize. It describes a person’s answers across questions or situations, not a recording of what they said after one identifiable mistake. The distinction matters. A measure of general readiness can reveal patterns across a group, but it cannot establish why a particular person apologized, whether the apology adequately addressed the harm, or whether the person meant every word. In “Please accept my sincere and humble apologies: The HEXACO model of personality and the proclivity to apologize,” researchers examined associations in two adult samples. They used self-reports and knowledgeable-observer reports of personality. The strongest reported association with apology proclivity was honesty-humility, whether personality was rated by the participants themselves or by someone who knew them. Self-reported conscientiousness was also positively associated in both samples. Agreeableness was a positive correlate in one sample only. These are correlations: they show that scores tended to vary together in those samples, not that a trait caused someone to apologize or that a person higher on a trait will apologize in a particular incident. The abstract does not provide a basis for turning those group-level associations into an individual forecast. The wording of the finding also sets a boundary. “Honesty-humility” names a dimension in the HEXACO personality model; it is not a moral certificate, and the result does not say that people who score higher are always honest or that an apology proves honesty. Likewise, an association involving conscientiousness does not mean that every conscientious person promptly takes responsibility. A trait measure summarizes tendencies across responses. An apology after a mistake depends on the act, the relationship, the circumstances, and the choices made in that moment. The study measured a reported disposition to apologize for transgressions, not the truth of an apology’s explanation or its effect on the person receiving it. An earlier measure illustrates the same distinction. “The disposition to apologize” reports four studies correlating responses from more than 900 undergraduate students to a newly devised Proclivity to Apologize Measure with other questionnaires. The study examined how scores on that measure related to reported constructs such as well-being, acceptance, seeking forgiveness, and agreeableness. Its participant group and correlational questionnaire design make the result useful as evidence that researchers have studied individual differences in reported willingness. They do not make it a population norm, a test of what someone will do after this mistake, or proof of what motivates an apology. Undergraduates answering measures are not equivalent to people observed navigating every kind of real conflict. So, if a personality profile suggests a broad tendency, use it to form a question about repeated behavior: across comparable mistakes, does this person usually acknowledge their part and try to make amends? Then look at the events and exceptions themselves. A single “sorry,” or its absence, cannot confirm a trait explanation. The association gives a lens for noticing a pattern; it does not tell you what happened inside someone’s mind in this one exchange.

Sources: Please accept my sincere and humble apologies: The HEXACO model of personality and propensity to apologize; The disposition to apologize

What does observed behavior add to self-reported tendency?

Self-reported willingness and observed apologizing answer different questions. A questionnaire asks how someone generally sees their own readiness to apologize; a coded interaction records what they did in one designed situation. The distinction matters because a person can endorse a general tendency without displaying it in every event, and a single observed response cannot establish a stable personality pattern. The study titled “Guilt, shame, and apologizing behavior: A laboratory study” was designed to add that limited behavioral measure to earlier work based on reports of apologizing. The researchers first measured dispositional guilt and shame among 59 undergraduates recruited from lower-level psychology courses at a small liberal arts college in the Midwestern United States. Participants were led to believe they had arrived late for the study. Two observers then coded how extensively each participant apologized. This created an opportunity to observe words and actions in response to an apparent, low-stakes mistake, rather than asking participants to describe their usual behavior. It did not observe an apology to someone they had actually harmed, or follow what happened in a relationship afterward. The study treated guilt and shame as separate measured tendencies, with distinct facets. Here, guilt-negative-behavior-evaluation means a tendency to judge a particular action negatively; guilt-repair means a tendency to want to make amends after wrongdoing. Shame-negative-self-evaluation concerns judging the self negatively, while shame-withdraw concerns wanting to withdraw from a shame-inducing situation. These are scale dimensions, not labels an observer can reliably assign from a person's face, silence, or choice of words. The researchers measured these tendencies before the staged lateness event; they did not establish what participants felt at that moment. The reported associations were not identical across facets. More extensive coded apologizing was positively correlated with guilt-repair (r = .46, p < .001) and guilt-negative-behavior-evaluation (r = .36, p = .005). The association with shame-negative-self-evaluation was positive but described as marginal (r = .23, p = .08); apologizing was not significantly related to shame-withdraw (r = −.03, p = .83). In plain terms, participants who scored higher on the two measured guilt facets tended to apologize more extensively in this particular setup. The result for shame does not support a simple claim that shame-prone people necessarily retreat instead of apologizing. A marginal association is uncertain evidence, and a nonsignificant result does not prove that no relationship exists. This behavioral result complements, but does not replace, research on self-reported proclivity to apologize. The measures differ: one captures a reported general readiness, while this study counts coded behavior during one induced event. Their convergence around guilt-related measures is compatible with a link between those tendencies and apologizing, but this small correlational study cannot show that guilt causes an apology. Nor can it tell whether an apology was sincere, adequate, or helpful to a recipient. For a specific mistake, the observation is therefore modest: behavior in one event gives more direct evidence of what happened there than a questionnaire about usual willingness, but it still leaves motive and broader pattern unresolved. An apparent delay before apologizing, or an apology itself, cannot reveal a private emotional state. What the study adds is a reason to keep measured tendencies, observed conduct, and inferred motive separate.

Sources: Guilt, shame, and apologizing behavior: A laboratory study

How is an explanation different from an excuse in this evidence?

An explanation answers a causal question: what contributed to the mistake? An apology addresses responsibility: what part of the act was wrong, and what does the speaker acknowledge? These can appear together. In the workplace study, an excuse is distinguished by shifting perceived responsibility toward an external or mitigating cause. “To Excuse or Not to Excuse: Effect of Explanation Type and Provision on Reactions to a Workplace Behavioral Transgression” defines an apology as expressing sorrow, acknowledging the act’s wrongfulness, and accepting responsibility. An excuse may reduce perceived responsibility by identifying an external, uncontrollable, or mitigating cause. A justification accepts responsibility but argues the act was warranted by a competing purpose. These are research categories; ordinary statements can combine them. Consider a missed promised call. “The train stopped unexpectedly; I should have let you know, and I’m sorry you waited” gives context while acknowledging a failure to communicate. “The train stopped, so you can’t blame me” uses the circumstance to argue against responsibility. This is an illustration, not a tested case. The study examined lateness to workplace meetings. In Study 1, 558 employed adults who regularly attended meetings viewed one of 12 video scenarios. Researchers varied whether the late arrival apologized, offered an excuse, or said nothing; whether the waiting group complained; and whether the person was habitually late. Participants rated expected work performance and willingness to help. An excuse produced more favorable performance ratings and greater stated willingness to help than silence. The apology was not reliably better than silence in the primary analyses; its performance comparison was marginal, and changed when people excluded for failed attention checks were included. The excuse received somewhat higher expected-performance ratings than the apology, a small difference. The abstract cautions that the apology and excuse distinction was not always clear. This finding is not advice to excuse lateness. The authors discuss different possible routes: an excuse may shape how a listener assigns cause and responsibility, while an apology acknowledges responsibility and may affect the response to it. Outcomes were judgments and intentions in a constrained meeting scenario, not later cooperation, restored trust, or sincerity. Group complaining and a history of habitual lateness also mattered, suggesting reactions depend partly on setting and prior pattern. For a specific mistake, ask: what does this account clarify, and what responsibility remains acknowledged? An obstacle can explain why a call was missed; it does not erase the effect of waiting or settle whether a message could have been sent. A response can name the act, acknowledge its impact, state the speaker’s part, then give context without treating it as proof that no repair is needed. The appropriate repair depends on the consequence and the other person’s boundaries; this study did not test a universal script. Nor can one explanation settle a personality question. Someone may give context to help the listener understand, reduce blame, or both. This study does not show that personality causes excuse-making or that an apology reveals a stable trait. To reflect on a tendency, compare similar events: does the person name their part, do explanations repeatedly replace acknowledgment, and does later conduct address the same problem? Such observations can support a tentative account of a recurring response, but cannot certify sincerity. Begin with the concrete act and its effect, then distinguish context from responsibility-shifting. That keeps the discussion specific without turning one mistake into a character verdict.

Sources: To Excuse or Not to Excuse: Effect of Explanation Type and Provision on Reactions to a Workplace Behavioral Transgression

Which parts of an apology are judged useful, and what does that not prove?

In two experiments, participants rated apologies with more components as more effective, and responsibility acknowledgment and repair were among the favored elements. These findings describe what people may value after a trust violation. They do not establish that a speaker feels remorse, that a recipient will forgive, or that trust will recover. The studies measured reactions to statements, not private sincerity or later conduct.

“An Exploration of the Structure of Effective Apologies” reports two experiments about a job candidate’s past tax-return error. In the first, 333 U.S. adults recruited through MTurk evaluated apology statements. One scenario described insufficient tax-code knowledge; another described knowingly filing an incorrect return. Six-component apologies received higher average evaluations than one- or three-component statements. Competence-error apologies also received higher ratings than integrity-breach apologies.

The six components were regret, explanation, responsibility acknowledgment, repentance, repair offer, and request for forgiveness. In the second experiment, 422 participants rated statements containing one to six components. Shown alone, repair and repentance received the highest average ratings, followed by responsibility acknowledgment; a forgiveness request alone received the lowest. Rankings varied across analyses, so the study does not supply a universal script or fixed order.

The scenario contrast shows why wording cannot be separated from the event. A competence-based violation involved limited knowledge; the integrity version involved a knowing choice. These describe the study’s cases, not personality categories or full explanations of real conduct. Participants judged competence-case apologies more favorably. Perceived circumstances may shape evaluations, but this does not mean a deliberate breach cannot be addressed or an accidental error needs no accountability.

Responsibility acknowledgment and explanation answer different questions: what part do I own, and how did this happen? If someone misses a promised call, saying the workday ran late gives context. Naming the unkept promise acknowledges the commitment. This example illustrates the distinction; it does not predict a recipient’s reaction. An explanation can be relevant without replacing responsibility.

An offer of repair matters when a consequence can be addressed. It is an effort to rebuild trust, when repair is possible. Its discussion distinguishes a verbal commitment from carrying it out: follow-through matters for credibility. After a missed call, providing needed information may help if it remains useful. That cannot recover lost time, and the recipient may prefer no further contact.

These ratings are not a test of sincerity or forecast of forgiveness. Participants imagined a candidate’s conduct in a hiring decision; they were not directly harmed, and the relationship was transactional, not close. Components appeared as words on a screen, without tone or emotional intensity. The measured outcomes were scenario evaluations, not later trust or behavior.

For a specific mistake, compare the apology with the event: does it name the action, acknowledge responsibility, and offer a feasible repair the affected person wants? This is an editorial application, not a validated checklist. Sometimes repair cannot undo harm; an offer can cross a boundary. A request for forgiveness was studied, but asking does not oblige the recipient to forgive or reassure the speaker.

The bounded conclusion is that in constructed trust-violation scenarios, fuller apologies tended to receive better evaluations, with responsibility and repair among the more favorably rated components. This cannot establish sincerity or predict relationship recovery. Event details offer more than a personality label: what happened, its impact, acknowledged responsibility, and repair. A trait interpretation requires a pattern across comparable situations and remains tentative.

Sources: An Exploration of the Structure of Effective Apologies; An Exploration of the Structure of Effective Apologies

Why can the same words mean different things in different situations?

The mistake's type, consequence, relationship, history, and possibility of repair shape what an apology can reasonably address. Research on workplace lateness and constructed trust violations can help distinguish questions to ask, but cannot supply a universal script. A useful comparison is between a competence-based failure, where someone lacked the knowledge or ability to meet an expectation, and an integrity-based violation, where the concern is a choice or breach of honesty. “An Exploration of the Structure of Effective Apologies” reports two studies in which participants evaluated apology components in trust-violation scenarios. Apologies with more components were rated more effective than those with fewer, and apologies after competence-based violations received more favorable ratings than those after integrity-based violations. These are evaluations of constructed scenarios, not observations that a particular apology repaired trust over time. The setting matters too. “To Excuse or Not to Excuse: Effect of Explanation Type and Provision on Reactions to a Workplace Behavioral Transgression” examined people arriving late to meetings. Across two studies using experimental and survey methods, an excuse was better than no explanation on measured reactions, while differences between apology and no explanation, and apology and excuse, were not always clear. Here, an apology includes sorrow, acknowledgment that the act was wrong, and acceptance of responsibility. An excuse offers an account that can reduce perceived personal responsibility; a justification accepts responsibility while arguing the behavior was warranted. This workplace context does not establish that one form will have the same effect elsewhere. Consider an illustrative case: someone forgets a promised phone call. If it is a one-time lapse, acknowledging it and proposing a realistic time to call may address the immediate inconvenience. If the person has repeatedly missed agreed calls, the recipient may be concerned about reliability and the pattern's effect, not only this instance. Repeating “I’m sorry” may express regret but does not by itself answer what will change. A real constraint may explain how the lapse happened without erasing its impact or the responsibility to communicate when possible. “Apologies in Close Relationships: A Review of Theory and Research” supports keeping the recipient and relationship in view. Its reviewed studies found that recipient and transgressor characteristics, along with context such as relational closeness, influenced whether apologies occurred and how they were evaluated. It supports attention to context, not a formula for predicting one person's response. As a practical inference from these sources, a bounded rule is: match acknowledgment and proposed repair to the act, its consequence, and the history between the people. This is a context-bound inference, not a universal finding. A workplace-lateness finding cannot dictate what to say after a broken confidence; a trust-scenario rating cannot establish forgiveness in a close relationship. Describe what happened, take responsibility for your part, and offer a feasible repair. When a mistake recurs, follow-through becomes relevant evidence about the pattern, though it cannot reveal private motive with certainty. Personality may help someone reflect on recurring tendencies across comparable situations, but these findings do not justify letting a label excuse a mistake or determine whether an apology is sincere.

Sources: An Exploration of the Structure of Effective Apologies; To Excuse or Not to Excuse: Effect of Explanation Type and Provision on Reactions to a Workplace Behavioral Transgression; Apologies in Close Relationships: A Review of Theory and Research

Can a belief that people change alter how an apology is read?

A belief that personality can change may affect how a recipient interprets an apology, but it cannot establish that the apologizer has changed. In “Apologies as signals for change? Implicit theories of personality and reactions to apologies during the #MeToo movement,” participants who more strongly endorsed personality change evaluated the apologies and the people who issued them more favorably, on average. The study measured recipients’ judgments of real public statements; it did not track whether the accused figures later changed their conduct. For a private mistake, this is evidence about one possible lens on an apology, not a guide to its sincerity. The study calls beliefs about whether personality is changeable implicit theories of personality. People leaning toward an “incremental” view see traits as more malleable; those leaning toward an “entity” view see them as more fixed. These are positions along a measured dimension, not permanent categories. The research used an online, nationally representative U.S. sample of 720 adults recruited through Qualtrics in December 2017. Participants read four real apologies from prominent men responding to sexual-misconduct allegations during the #MeToo movement and rated the statements and apologizers. More incremental beliefs were associated with more favorable ratings across several outcomes, including perceived apology quality, evaluations of the apologizers, forgiveness, and lower punitiveness. The overall association with positive evaluations was modest: adding personality-change beliefs reduced unexplained variance by 4 percent in the study’s model. Participants with stronger incremental beliefs were also more likely to say redemption was possible. These are associations between recipients’ beliefs and their judgments, not evidence that one belief produces fairer judgments or that redemption occurred. The authors’ proposed interpretation is that a recipient who believes people can change may read an apology as a signal that the person is willing to improve. Someone with a more fixed view may give the past conduct more weight when anticipating future behavior and regard the apology as less informative. This offers two plausible ways of reading the same words. It does not tell us which reading is correct in a particular case. The context also sets a sharp boundary. These were public apologies concerning high-profile allegations, read alongside information about the allegations. The statements varied in comprehensiveness and defensiveness, and the authors note that the sample of four may not represent the wider set of #MeToo apologies. Public statements to a broad audience differ from a private apology after, for example, forgetting a promised call. The study does not show that recipient optimism is evidence of changed behavior, nor that skepticism is proof that a person cannot change. It also cannot settle the response of someone directly harmed: the participants were evaluating public apologies, not deciding how a particular harmed person should feel. For an everyday mistake, separate your view of human change from the evidence in front of you. Ask what the apology names, whether responsibility is clear, and what happens afterward. A recipient may reasonably remain uncertain; an apologizer’s promise is not the same as follow-through. Likewise, a history of similar conduct matters to judging whether a response addresses the concern, but it does not let an observer read private motives with certainty. The useful lesson is modest: beliefs about change can color how an apology is received. Treat that as a reason to check your interpretation, not as a verdict on the apology or the person.

Sources: Apologies as signals for change? Implicit theories of personality and reactions to apologies during the #MeToo movement

How should trait language stay tentative?

A useful reflection starts with an event description that another person could recognize: what was said or done, when it happened, and what consequence followed. “I am careless” already interprets the event; “I said I would call after work and did not call” describes it. Begin there because the research on apology tendency concerns broad reported patterns, not a reliable reading of a single incident. “Please accept my sincere and humble apologies: The HEXACO model of personality and the proclivity to apologize” associated personality measures with general reported readiness to apologize across two adult samples. This can prompt a question about repetition, but cannot explain one apology or establish its adequacy. Next record context that could change what the act means: the stakes, what the person knew, constraints at the time, the relationship, and whether the same commitment had been missed before. Context does not erase responsibility. A one-time failure to call during an unexpected disruption differs from repeatedly making promises without checking whether they can be kept. This is a reflection aid, not a tested personality measure. Then describe the response in observable terms. Did the person name the action? Acknowledge its effect? Accept responsibility for their part? Offer a feasible repair or follow through later? Record what was actually said and done, without turning a quiet voice, quick reply, tears, or lack of visible emotion into proof of remorse or its absence. The studies measured reported tendencies, coded behavior in one staged event, apology-content judgments, or recipient evaluations in a public context. None establishes an outward cue for private sincerity. Words are observable; motive is less directly available. Only after setting out event, context, and response is recurrence useful. Compare similar moments and exceptions: Do they acknowledge an error when it is pointed out, explain circumstances before recognizing their part, or complete a promised repair? When do they respond differently? A missed call, broken confidence, and public insult differ in stakes; combining them into one measure of “how apologetic someone is” hides important distinctions. From that record, form a tentative interpretation, not a character verdict. You might say, “In these repeated low-stakes commitments, I have seen delayed acknowledgment and little follow-through; I want to ask what gets in the way.” You could also find that the person apologized promptly and repaired the specific impact, while the event remains upsetting. Neither observation licenses a fixed label. “Apologies as signals for change? Implicit theories of personality and reactions to apologies during the #MeToo movement” found that recipients’ beliefs about personality change were associated with their evaluations of real public apologies. Its 720-person U.S. study concerned high-profile allegations and public statements; recipient optimism or skepticism is not evidence that an apologizer changed. Let a personality idea generate a question, then update it with comparable events, exceptions, and later conduct. Revise it when new behavior conflicts. That keeps self-understanding useful without turning one mistake into diagnosis, excuse, condemnation, or a promise that change has occurred.

Sources: Please accept my sincere and humble apologies: The HEXACO model of personality and propensity to apologize; Apologies as signals for change? Implicit theories of personality and reactions to apologies during the #MeToo movement

What is the next useful step after the mistake?

Begin with the event, not a personality label. Name what you did, describe its consequence, and acknowledge your part. Ask what would help now and whether repair is possible. After a missed call, for example: “I said I would call and didn’t. I’m sorry. Would another time help, or would you rather leave it for now?” This is an illustration, not a formula. The other person may want space, a different remedy, or no further conversation. An apology does not require forgiveness.

Context can matter without swallowing the acknowledgment. “To Excuse or Not to Excuse” reports two studies of workplace meeting lateness: an excuse was better than no explanation, while comparisons involving apology were not always clear. That narrow setting cannot prescribe how to respond to a missed promise or serious harm. Its useful prompt is whether an account clarifies what happened while leaving responsibility visible, or asks the other person to discount the impact.

Afterward, look for follow-through rather than treating wording as proof of change. “An Exploration of the Structure of Effective Apologies” reports that participants rated apologies with more components as more effective in trust-violation scenarios. Those judgments do not establish what one recipient will accept or whether trust will recover. For self-reflection, compare similar events: what was acknowledged, repaired, repeated, or different? This may suggest a tentative tendency, not reveal private sincerity. Ask what the other person needs now, accept their answer, and make only a repair you can sustain.

Sources: An Exploration of the Structure of Effective Apologies; To Excuse or Not to Excuse: Effect of Explanation Type and Provision on Reactions to a Workplace Behavioral Transgression

Questions readers ask